Mailing-List: contact dev-help@cocoon.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@cocoon.apache.org
Received-SPF: pass (asf.osuosl.org: local policy)
Mime-Version: 1.0 (Apple Message framework v622)
In-Reply-To: <42B92C96.4080309@apache.org>
References: <d0bbddaacd5a1f854fc55579bd785541@wrinkledog.com>
 <42B7B6CF.8040701@apache.org>
 <21a822398d88ce3b949150b1084c1b93@wrinkledog.com>
 <42B92C96.4080309@apache.org>
Content-Type: multipart/mixed; boundary=Apple-Mail-9--481528772
Message-Id: <d632a305ad0bfff6b115801638ec9d70@wrinkledog.com>
From: Mark Lundquist <ml@wrinkledog.com>
Subject: Re: [CForms binding] access to model data from repeater row widget
Date: Thu, 23 Jun 2005 13:46:31 -0700
To: dev@cocoon.apache.org


--Apple-Mail-9--481528772
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=WINDOWS-1252;
	format=flowed


Hi Sylvain, thx for your reply...

>  ....
>>         However...
>>
>>>>         I realized that I have a nomenclature clash with my choice=20=

>>>> of
>>>>         the name "model". I've been using the v2 API for a long=20
>>>> time,
>>>>         so I forgot in v1, there is a Form.model property that=20
>>>> denotes
>>>>         the widget value tree.
>>>
>>>     The "model" property is only available on the toplevel form JS
>>>     object, so there may not be a clash.
>>
>> It's not a hard "name clash", but it is a nomenclature clash in that=20=

>> it's a different sense of the term "model". My new row "model"=20
>> property is truly the "model", as in "MVC" or as in=20
>> "loadFormFromModel()". The v1 "model" is "model" in terms of=20
>> "formmodel", I guess. I just think it's too confusing to have both=20
>> senses. I actually sort of wish the v1 "model" were called something=20=

>> else, because it isn't the model /a la/ MVC.
>
> You're right. The "form.model" is actually the value tree.

Maybe form.model should be renamed to "form.value"!  Then you would=20
have these two expressions that mean the same thing (just as they look=20=

like they should! :-):

	form.value.foobar

and

	form.lookupWidget('foobar').value

, right? :-)

But for now in my working version of the code I settled on "rowModel"=20
to expose the row model object in Flowscript and JXT.

Now then, the following exchange has me totally confused:

>>>     It doesn't seem good to me to provide at the API level a way to
>>>     link the form to the data model,
>>
>> No, no, it /is/ good... :-)
>> [ ...my rambling snipped...]
>>
>>>     However, there are some uses cases where this makes sense. You=20=

>>> can
>>>     then use widget attributes (see get/setAttribute) to attach some
>>>     application data of your liking to any widget, including =
repeater
>>>     rows.
>>
>> See, that's the problem, this is what I have to deal with currently=20=

>> :-)... I have two parallel structures: a Collection, and a Repeater.=20=

>> If the Repeater does not give me a "view" to its collection, then I=20=

>> have to do some kind of iteration to create associations somewhere=20
>> between the repeater rows and the collection elements, and this just=20=

>> feels silly and wasteful. I've done it this way, decorating the=20
>> repeater rows with handles to the model data, and I just think that=20=

>> since the binding layer is already iterating across the collection,=20=

>> it should do this decoration for me! :-).
>
> Yes. That's the idea behind the "bound repeater binding" idea Reinhard=20=

> pointed to.

Right.  Which was your idea, and which I think I have gone and=20
implemented before I knew that somebody else had thought of it :-)... =20=

Or have I not?  Maybe I am misunderstanding Reinhard's reference and it=20=

isn't really the same thing... or maybe you are misunderstanding me,=20
and thinking it's not the same thing, but it really _is_? :-)

Are you saying, "Ah yes, you're right... I thought of this before, and=20=

now I remember why it _is_ a good idea (+1! :-)" ?

Or =97 are you saying "Ah, it looks like I did think of this too once,=20=

but now on second thought I'm not so sure it was a good idea" ?

>> I could live with there being a "callback" facility in the repeater=20=

>> binding language, e.g. <fb:javascript> within <fb:on-bind>... know=20
>> what I mean? But that's sort of beyond my ability to implement, and I=20=

>> was looking for something that I could actually do :-)
>
> Extending the repeater binding to keep the connection to the data=20
> model may be easier, no?

Well, it certainly was awfully easy :-)  Tell you what, the change is=20
so ridiculously small, I'll just attach the diffs  (they are against=20
src/blocks/forms/java/org/apache/cocoon/forms) and you can see if we're=20=

talking about the same thing.  At first this amounted to about 6 lines=20=

of code added, before I added the convenience flowscript function (that=20=

I had mentioned in the original post that I wanted to add... this=20
doubled :-) the size of the mod).  BTW, this function =97 =
'getRowModel()'=20
=97 I only added it in v2 ('cuz that's what I'm using! :-).  When I=20
looked at v1's ScriptableWidget, I hesitated when I saw that there are=20=

so few javascript functions defined there, and I wasn't sure if this is=20=

really the right idiom for v1... Care to comment?

Cheers,
=97ml=97

--Apple-Mail-9--481528772
Content-Transfer-Encoding: 7bit
Content-Type: application/octet-stream;
	x-unix-mode=0644;
	name="rowModel.diff"
Content-Disposition: attachment;
	filename=rowModel.diff

Index: formmodel/Repeater.java
===================================================================
--- formmodel/Repeater.java	(revision 594)
+++ formmodel/Repeater.java	(working copy)
@@ -311,6 +311,11 @@
 
     public class RepeaterRow extends AbstractContainerWidget {
 
+        private Object model;
+
+        public Object getModel() { return model; }
+        public void setModel (Object model) { this.model = model; };
+
         public RepeaterRow(RepeaterDefinition definition) {
             super(definition);
             setParent(Repeater.this);
Index: generation/template.jx
===================================================================
--- generation/template.jx	(revision 594)
+++ generation/template.jx	(working copy)
@@ -63,6 +63,7 @@
   <jx:set var="lastRow_" value="${repeater_.length - 1}"/>
   <jx:forEach varStatus="loop_" begin="0" end="${lastRow_}">
       <jx:set var="context_widget_" value="${repeater_.getRow(loop_.index)}"/>
+			<jx:set var="rowModel_" value="${context_widget_.unwrap().model}"/>
       <jx:evalBody/>
   </jx:forEach>
 </jx:macro>
Index: flow/javascript/v2/ScriptableWidget.java
===================================================================
--- flow/javascript/v2/ScriptableWidget.java	(revision 594)
+++ flow/javascript/v2/ScriptableWidget.java	(working copy)
@@ -682,6 +682,18 @@
         }
     }
 
+    public Object jsFunction_getRowModel()
+        throws JavaScriptException
+    {
+        if (delegate instanceof Repeater.RepeaterRow) {
+            return ((Repeater.RepeaterRow) delegate).getModel();
+        } else {
+            throw new JavaScriptException (
+                            "expected a Repeater.RepeaterRow instead of " + delegate
+                        );
+        }
+    }
+
     static final Object[] WIDGET_CLASS_MAP = {
         Form.class, "Form",
         Field.class, "Field",
Index: binding/RepeaterJXPathBinding.java
===================================================================
--- binding/RepeaterJXPathBinding.java	(revision 594)
+++ binding/RepeaterJXPathBinding.java	(working copy)
@@ -121,6 +121,7 @@
             // make a jxpath ObjectModelSubcontext on the iterated element
             Pointer jxp = (Pointer)rowPointers.next();
             JXPathContext rowContext = repeaterContext.getRelativeContext(jxp);
+            thisRow.setModel (rowContext.getValue ("."));
             // hand it over to children
             this.identityBinding.loadFormFromModel(thisRow, rowContext);
             this.rowBinding.loadFormFromModel(thisRow, rowContext);

--Apple-Mail-9--481528772--